Ensemble Based Learning Style Identification using VARK

Abstract

 In the current situation the need for e –learning grown larger in both education and training industry. The scope of e-learning paves way to adapt Learning Management System (LMS) an integrated web based learning environment and tool for instructional purpose to provide learning available at any time anywhere to the learner becomes essential tool for learners.  Learning is considered as legacy process that differs for every individual. Each individual has their own Learning style in adapting new and concrete information. Each learning style got it own  individual method  in understanding learners learning style, among these VARK model developed by Fleming is widely accepted to enhance its functionality with the recent technologies. In this work we used ensemble learning a machine learning Meta approach to gain better predictive performance by aggregating the predictions from multiple models. Our main objective of adapting ensemble approach in learning prediction using VARK model has been carried out using classifiers such as J48, SVM, Naive Bayes and Random forest as initial step towards our objective. The bagging ensemble approach has been utilized under hard majority voting to improve more accuracy in learning style identification and to identify various attributes that influence in personalizing LMS. Thus, the efficient personalization of learning management system by understanding learners learning style helps to provide sophisticated Learning Environment to the Learner. It can also be extended in various levels of training in the corporate industry for the effective management of employee training and learning process can be made easier to the learner.